We wish to bring to full development a new technique for recognition and localization of all or nearly all single base changes in any selected region of human genomic DNA, relative to wild-type or any reference sequence. The technique is based on recent work in this laboratory, together with earlier advances in denaturing gradient electrophoresis. Although the procedure requires cloned probes of the genomic region to be scrutinized, it does not require that the base sequence be determined, and it requires little time and effort. We wish to carry out complete scrutiny of the human beta globin gene, with challenge by a large number of genomic samples carrying transcription or coding defects that have been identified in the gene sequence. This will require fewer than one dozen probes. It will then be appropriate to survey of a small population to determine the presence of polymorphisms or idiosyncrasies of genetic variation that might be confused with significant sequence changes. We will work toward comprehensive scrutiny of much larger genes. Of particular interest are hemophilia A - the factor VIII gene, activation of the ras oncogene, chronic granulomatous disease, and Alzheimer's Disease. We plan to bring to full development a new gradient instrument that will both simplify the denaturing gradient procedure and will improve the quality of the data substantially. We plan to explore certain aspects of DNA melting, on which this system depends, and changes in the sequence-dependence of melting that can be induced by different environments to provide an alternative approach for the comprehensive detection of all sequence changes. We will continue thermodynamic studies on the effects on thermal stability of non-Watson-Crick defects in the DNA helix. This work is expected to provide an effective, simple, and widely applicable means for the diagnosis of genetic disease. It will contribute to genetic diagnosis by direct recognition of defects in the genes of an individual who may or may not already display physiological signs of the problem, or in prospective parents.